Tracing Dynamical Evolution of Dark Matter via Machine Learning

We plan to answer two pivotal questions of modern astrophysics: the nature of dark matter and its interaction with baryonic processes.

Utilizing galaxy observations and cosmological hydrodynamical galaxy simulations across a redshift range of z = 0.3-2.5, we will examine 3-10 Gyr of cosmic history.

We propose to ""Trace the Dynamical Evolution of Dark Matter via Machine Learning""- TraDE-DML, that pioneers an advanced methodology for assessing the dynamical masses of galaxies, aiming for unprecedented precision in the quantification of both baryonic and dark matter components.

Unlike conventional velocity profile studies, TraDE-DML eliminates assumptions of symmetry and dynamical equilibrium, substantially reducing uncertainties in dark matter estimates.

Our project aims to exploit existing and future survey data, preparing for expansive telescopic projects like ELT and SKA.

Simple in concept but revolutionary in application, the machine learning techniques used in TraDE-DML are poised for transformative advances in dark matter studies, particularly in determining its central density slope.

By synergistically integrating knowledge from observational astronomy, theoretical physics, machine learning, and statistics, TraDE-DML aims to make significant strides in unraveling the elusive nature of dark matter.

As an expert in observational data analysis with privileged access to leading galaxy surveys like MAGPI and MIGHTEE, I possess the skills to efficiently extract and analyse pertinent data. The host, Dr. Benoit Famaey, excels in galaxy dynamics and alternative dark matter theories.

Supported by a team versed in cosmological simulations and machine learning experts at the Inter-disciplinary Institute IRMIA++, we form a unique research synergy.

Utilizing advanced machine learning frameworks and leveraging expansive survey data, TraDE-DML is well-positioned for immediate execution.